Analysis of specific serum markers of colon carcinoma using a Bhattacharyya-based support vector machine

Abstract

We aimed to evaluate the specificity of 12 tumor markers related to colon carcinoma and identify the most sensitive index. Bhattacharyya distance was used to evaluate the index. Then, different index combinations were used to establish a support vector machine (SVM) diagnosis model of malignant colon carcinoma. The accuracy of the model was checked. High accuracy was assumed to indicate the high specificity of the index. The Bhattacharyya distances of carcinoembryonic antigen, neuron-specific enolase, alpha-feto protein, and CA724 were the largest, and those of CYFRA21-?, CA125, and UGT1A83 were the second largest. The specificity of the combination of the above seven indexes was higher than that of other combinations, and the accuracy of the established SVM identification model was high. Using Bhattacharyya distance detection and establishing an SVM model based on different serum marker combinations can increase diagnostic accuracy, providing a theoretical basis for application of mathematical models in cancer diagnosis.